![Mean shell cover (%, ± 1 standard error [SE]) within the nest territories of Australian Fairy Tern, Sternula nereis nereis, at a colony in North Fremantle, Western Australia. Nest establishment = days after the first nest was established.](https://www.researchgate.net/profile/Claire_Greenwell3/publication/344506826/figure/tbl1/AS:973930445668357@1609214521913/Mean-shell-cover-1-standard-error-SE-within-the-nest-territories-of-Australian_Q320.jpg)
Content uploaded by Claire Greenwell
Author content
All content in this area was uploaded by Claire Greenwell on Feb 06, 2021
Content may be subject to copyright.
This article has been accepted for publication and undergone full peer review but has not been through the
copyediting, typesetting, pagination and proofreading process, which may lead to differences between this
version and the published article. Please cite this article as https://doi.org/10.1071/PC20056
This article is protected by copyright. All rights reserved.
The secret life of Fairy Terns: Breeding chronology and life history observations of
Sternula nereis nereis in south-western Australia
†Greenwell, C.N.1,2, Dunlop, J.N.1,3, Admiraal, R.4, & Loneragan, N.R.1,2
1Environmental and Conservation Sciences, College of SHEE, Murdoch University, 90 South Street,
Murdoch, Western Australia, 6150, Australia
2Centre for Sustainable Aquatic Ecosystems, Harry Butler Institute, Murdoch University, 90 South
Street, Murdoch, Western Australia, 6150, Australia
3Conservation Council of Western Australia, Lotteries West House, 2 Delhi Street, West Perth,
Western Australia, 6150, Australia.
4Victoria University of Wellington, School of Mathematics and Statistics, Kelburn Parade, Wellington,
6012, New Zealand.
†Corresponding Author: c.greenwell@murdoch.edu.au
Abstract
This research describes the breeding ecology, behaviour and substrate preferences of the
Australian Fairy Tern, Sternula nereis nereis, in four colonies around Perth, Western Australia
between 2018-2020. Extensive field observations, supported by a bird banding study and
sunrise to sunset video recording were used at colony and roosting sites to determine the
processes of mating, colony formation, egg-laying and incubation periods, post-hatching care
and breeding success (fledglings per pair). At a colony in North Fremantle, the median nest
spacing was 0.71 m (mean 1 SE = 0.89 0.05 m), which increased over time. Birds
establishing nests within a week of the first eggs being laid selected sites with significantly
higher percentage beach shell cover (73.5% 4.5%) than those laying later in the season
(58.2% 7.9%) and on average, birds selected sites with higher shell cover (64.9% 2.8%, n
= 114) than a random sample of sites within the colony (53.7% 4.4%, n = 44). Incubation
periods ranged from 17-26 days (n = 86, mean = 21 0.17 days). Incubation shift duration was
highly variable, with both sexes contributing, almost equally, to the care of the brood (mean =
1.27 ± 6.11 hr). Chicks fledged 21 – 23 (mean = 22 0.21, n = 10) days following hatching,
with all banded juveniles leaving the colony site within 8 days of fledgling. The information
gained from this research informs conservation strategies for this vulnerable species, where
management interventions are frequently necessary to prevent population decline.
2
Introduction
The Fairy Tern, Sternula nereis, is one of seven small terns in the genus Sternula and
one of two found within Australasia. Three distinct and geographically isolated subspecies of
Fairy Tern are recognised: Sternula nereis davisae (New Zealand); Sternula nereis exsul (New
Caledonia); and Sternula nereis nereis (Australia) (BirdLife International 2018), with genetic
evidence suggesting that S. n. nereis in Western Australia may have founded populations in
eastern Australia, New Zealand and New Caledonia (Baling & Brunton 2005; Dunlop &
Greenwell 2020).
Fairy Terns are coastal seabirds that feed, largely, on small schooling baitfishes in
families such as the Atherinidae and Clupeidae (Johnstone & Storr 1998). They exploit a
variety of coastal breeding habitats, but most commonly select open, lightly coloured, coarse-
grained sand spits and beaches. These sites are often at or near the mouths of estuaries or on
nearshore islands, within sheltered coastal embayments or coastal salt lakes, near aggregations
of small baitfish. (Higgins & Davies 1996; Johnstone & Storr 1998; Dunlop 2018).
The breeding behaviour of Fairy Terns exposes them to a broad range of threats and,
consequently, populations have been decreasing over much of their range in recent decades
(Threatened Species Scientific Committee 2011; BirdLife International 2018). The Australian
Fairy Tern was listed as vulnerable under the Environment Protection and Biodiversity
Conservation Act (1999) in 2011 after a ~24% population decline was recorded between 1974
and 2007 (Department of the Environment 2011; Threatened Species Scientific Committee
2011). Yet, there have been no major studies examining their behaviour, breeding chronology
and life history characteristics in Australia, despite an urgent need for research and
conservation effort (Higgins & Davies 1996).
As one of New Zealand’s rarest birds (≤ 10 breeding pairs), S. n. davisae has been a
subject of interest and intense management since 1991, with research concentrated on breeding
3
behaviour, genetics and ecology to inform species conservation (e.g. Baling & Brunton, 2005;
Ferreira et al., 2005; Jeffries & Brunton, 2001). However, low population size is likely to be
obscuring behaviour that might otherwise be observed in a larger breeding population. For
example, in Australia and New Caledonia, Fairy Terns are gregarious and typically nest in
relatively closely spaced colonies (Higgins & Davies 1996; Baling et al. 2009). In contrast, in
New Zealand, pairs nest in solitude and appear to defend large breeding territories against
conspecifics (Parrish & Pulham 1995; Jeffries & Brunton 2001), which may be a strategy to
minimise predation (Lack 1968; Perrins & Birkhead 1983) and to guard mates during the
breeding period, respectively. Therefore, the behavioural ecology of S. n. exsul and S. n. nereis
and the cues used in the selection of colony sites may differ from that of S. n. davisae.
In Australia, the distribution of the Fairy Tern in the west extends south from the
Dampier Archipelago to the Recherche Archipelago (Department of the Environment 2018).
Eastern populations are probably, largely, separated from those in the west by the Great
Australian Bight, with low numbers (< 1,600) found in South Australia, Victoria, New South
Wales, and Tasmania (Department of the Environment 2018), although, gene flow between
western and eastern Australia has been estimated at a rate of 5.26 migrants per generation
(Baling & Brunton 2005). Within Western Australia, there appear to be two distinct sub-
populations: a resident, winter-breeding (June - August) one restricted to the Pilbara region
(north from Exmouth Gulf), and a semi-migratory spring/summer one that breeds in colonies
along the coast south from Exmouth Gulf to the south-west capes and then east almost to the
WA border (Dunlop & Greenwell 2020). These birds breed between October and March. Seven
distinct areas, termed ‘neighbourhoods’ have been identified as conservation priorities for
S. nereis nereis in Western Australia (see Dunlop & Greenwell 2020).
This study describes the breeding chronology and aspects of the life history of the
Australian Fairy Tern (hereafter Fairy Tern) in south-western Australia. Extensive field
4
observations, combined with a bird-banding study and video analysis, provide detailed
accounts of mating behaviour, colony formation processes, substrate preferences, egg-laying
and incubation periods, time to fledging, post-hatching care and colony productivity (i.e.
number of fledglings produced per pair). The behaviours observed during this study have been
interpreted in light of a framework of Laridae breeding behaviour and ecology. The results
from this research should be used to help inform conservation strategies, where management
interventions are frequently necessary to improve breeding success in the key conservation
areas for this species, such as those recently identified by Dunlop and Greenwell (2020).
Methods
This study combines extensive field observations (≥ 250 days), video analysis (~226
hours) and bird-banding to describe the life-history of the Fairy Tern in south-western
Australia. These studies were undertaken within an 80 km radius of Perth, the major city and
capital of Western Australia (Fig. 1). A review of literature of the breeding behaviour and
ecology of species in the Laridae was undertaken to provide further insight and explanation of
field research findings. This research was conducted in accordance with Murdoch University
Animal Ethics approval (Permit R3077/18).
Arrival in south-western Australia
Surveys were undertaken on an ad hoc basis at sunset on Parkin Point, Garden Island,
Point Walter, Bicton (32° 00'40.32" S, 115° 47'11.76" E), and several sand spits on the Peel-
Harvey Estuary (Fig. 1) to confirm the presence of night roosts and the arrival of pre-breeding
birds during October and November of 2018 and 2019 and post-breeding aggregations in
February 2020.
Mating systems
Observations of mating behaviour, night roosting, colony formation and nesting
behaviour were undertaken at North Fremantle (32°02′25.83″S, 115°44′23.69″E), Point Walter
5
and Mandurah (32°31′14.24″S, 115°43′00.26″E, Fig. 1) during the 2018/19 season and on
Penguin Island, Shoalwater Bay (32°18′19″ S, 115°41′28″ E), where Fairy Terns were recorded
nesting for the first time in in 2019/20.
Figure 1. 'Locations of colony (■), roost (▲), and colony and roost (●) sites for Australian Fairy Tern, Sternula
nereis nereis, studied along the Western Australian coastline between 2018 and 2020. Δ denotes Bureau of
Meteorology weather station.
Colony formation, egg-laying and incubation
In 2018/19, the North Fremantle site was marked-out in a 5 m x 5 m grid, with plots
identified using 300 mm Jarrah survey pegs before the breeding season. Stakes were painted
white to the top one third and a black paint pen was used to label each peg in an alphanumeric
grid system. Pegs were then driven ~ 250 mm into the ground. During the breeding season,
observations of active nest sites were recorded at North Fremantle between 30 November 2018
and 14 January 2019, where a total of 220 nests were recorded. To determine the mean
incubation period, 92 nests were monitored from the date of egg laying through to hatching.
6
Birds attending active nest scrapes were observed from vantage points on the outskirts of the
colony using Swarovski EL 10x32 binoculars, and brooding behaviour, i.e. characteristic
settling movements that enable contact between the highly vascularised brood patch and the
egg, was used to assess for the presence of an egg. Where possible, observations into the nest
scrape were made to confirm egg presence, but in order to minimise colony disturbance, this
was not always attempted.
Newly established nests were recorded and mapped each day (usually before 9:00 and
after ~16:00), and when a second egg was observed within an established nest, the date was
recorded. Landmarks such as shells, rocks, vegetation and survey pegs were used to aid nest
mapping. A subset of nests containing two eggs (n = 19) were monitored daily to determine
the time between hatching in multiple egg clutches. The time between hatching in multiple egg
clutches was also recorded at Penguin Island (n = 8) during the 2019/20 season.
To confirm the presence of chicks, nests were monitored on a daily basis over several
hours in the early morning (before 09:00) and late afternoon (after 16:00). Often, the presence
of eggshell near nest sites aided in the detection of newly hatched chicks as shells are usually
removed by adults soon after hatching. On occasion, adults carrying eggshells away from the
colony (at any time of day), and the observation of wet chicks, enabled the exact time of
hatching to be determined. Where chicks were observed in the early morning, a combination
of appearance (i.e. wet or dry down feathers), behaviour (i.e. chick vigour and steadiness) and
the presence of eggshell within the nest, was used to estimate whether the chick had hatched
before- or post-midnight. If the chick was wet and/or very unsteady when presented with a fish,
or eggshell was still present within the nest, hatch date was recorded as post-midnight. If the
chick was dry and able to sit upright when presented with a fish, then the date of hatching was
recorded as hatched before midnight.
7
Shell cover preferences
An investigation into beach shell cover preference at North Fremantle was undertaken
by estimating the percentage of shell cover surrounding a subset of nests (n = 114), in a circle
with a diameter of 250 mm around each nest. A linear regression model of arcsine square root
transformed percentage shell cover on the date of nest establishment was undertaken to
determine whether there was a change in the average percentage shell cover surrounding nest
sites over time. To determine whether the average percentage shell cover at nest sites differed
from that of a random sample, 45 random points were plotted across the site and sampled during
the non-breeding period. Random x, y points were plotted onto a regular grid, using R (R Core
Development Team 2011) and the ‘raster’ (Hijmans & van Etten 2012) add-on package. One
point was excluded as the area was covered by a chick shelter. A t-test was used to determine
whether the percentage shell cover (arcsine square root transformed) differed between nest
sites and randomly sampled sites. Note that as the site is shell-rich, which holds sand in place,
and there is vegetation on the perimeter of the site to buffer from prevailing winds, there is
little sand movement, even during periods of strong winds. Therefore, shell cover was been
assumed to be static.
Parental roles
Parental roles, incubation routines and the period over which courtship feeding occurs,
were determined from sunrise to sunset video observations made at Mandurah (4 days) in
November 2018 and North Fremantle (10 days) in December 2018. Video recordings were
made of nests at various stages, i.e. before egg-laying and soon after laying through to hatching,
and a number of repeat measures were undertaken. Approximately six to eight nests were
captured within each frame during recording sessions. Eleven nests were selected for
observation, based on one member of the pair carrying a leg band or unique identifying feature
(e.g. black-tipped bill, scarring or black markings around the head). Courtship feeding and
8
copulations were used to confirm the sex of each bird within the pair. Repeat measures were
recorded at four nests. In total, 17 days of nest observations were made totalling > 242 h (Supp.
Table 1). Since it is possible that shift changes occurred before and after the sampling period,
the first and last observations for each day were excluded to remove uncertainty surrounding
the length of an incubation shift. By excluding these points, only observations for which the
actual incubation length was known (~ 192 h) were included in the analysis.
Incubation shift duration
To investigate the relationship between incubation shift duration and weather variables,
30-minute interval data was obtained from the Australian Bureau of Meteorology for average,
highest, and lowest wind speed and air temperature as well as total rainfall and relative
humidity. These data were obtained for the period of November and December 2018 for both
Mandurah and Swanbourne (~ 10 km north of Rous Head), respectively. Corresponding to each
incubation shift at a given site, average wind speeds, average air temperatures, total rainfall,
and relative humidity; maximum of highest wind speeds and air temperatures; and minimum
of lowest wind speeds and air temperatures for time intervals overlapping that shift for that site
were recorded. Two hundred repetitions of 10-fold cross-validation were performed for linear
regression models of log-transformed incubation shift duration on all possible combinations of
these variables to assess the best set of predictors of incubation shift duration using R (R Core
Development Team 2011) and the add-on packages of: ‘boot’(Canty & Ripley 2020), ‘car’
(Fox & Weisberg 2019), ‘doParallel’ (Microsoft Corporation & Weston 2019), ‘foreach’
(Microsoft Corporation & Weston 2020), ‘scales’ (Wickham & Seidel 2019) and ‘stringr’
(Wickham 2019). The simplest model, i.e. the model with fewest parameters, did not differ
significantly from the model with minimum mean squared error, which included as predictors
the highest and lowest wind speeds and air temperatures as well as relative humidity (average
root mean squared error of approximately 0.75).
9
Time to fledging
To determine time to fledging, 15 one- to four-day old chicks from single egg (5) and
double egg (5) clutches were hand captured at the nest scrape at the Penguin Island colony and
banded (Supp. 1, Table 3). The chicks selected for observation were representatives of the
colony as a whole at the time. Australian Bird and Bat Banding Scheme (ABBBS) bands
(painted with one of 15 nail-vanish colours) were placed on the right leg for unique
identification. Banding was conducted outside the colony and chicks were subsequently
returned to their nest scrape after banding. In all cases, adults immediately resumed brooding
once chicks were returned. Birds were followed on a daily basis, for at least 7 h per day until
fledging. For the purpose of this study, fledging is defined as the development of controlled
and independent flight, i.e. not wind assisted. The sight-recapture of banded juveniles away
from the Penguin Island colony by members of the WA Fairy Tern Network (for detail, see
Dunlop & Greenwell 2020) provided information on the timing of post-fledging dispersal.
Results and Discussion
Arrival in south-western Australia
Fairy Terns began arriving in the Perth metropolitan region in September where they
aggregated to forage near diurnal club sites before assembling at centralized night roost(s)
(Dunlop & Greenwell 2020). The number of individuals arriving in the south-west increased
considerably by the end of October, when larger flocks of Fairy Terns were observed at diurnal
club sites and/or night roosts (i.e. Perth to Mandurah, Fig. 1). Garden Island and the Peel-
Harvey Estuary (Creery Wetlands area) appeared to function as a night roost before the
commencement of breeding, where several hundred Fairy Terns arrived and departed during
twilight hours. The Point Walter islet was an important night roost, but the roosting of the birds
at this location occurred closer to the commencement of egg-laying than at other roost sites
10
(Dunlop & Greenwell 2020). Availability of sandbars and the high numbers of baitfish that
occur within these shallow water environments probably makes the sites highly attractive to
Fairy Terns.
Mating systems
Before the commencement of breeding, Fairy Terns undergo a moult and acquire full
nuptial plumage, i.e. a solid black head cap and a bright orange bill and legs, and arrest wave
moult of the primary feathers (see Dunlop 2018). This change in plumage state is associated
with gonadal recrudescence, i.e. a period of active gametogenesis, where the gonad and sex
steroid hormone concentrations increase in preparation for breeding (Foster 1975; Dunlop
1985a). On arrival in the south-west, many birds were well advanced in their moult, carrying
only a black tip at the end of their bill. For others, the acquisition of full nuptial plumage
occurred later in the season, sometimes several months after the first colonies were established.
Birds reaching peak reproductive condition later in the season were likely to be first-time or
less experienced breeders (Dunlop & Jenkins 1992). These changes in appearance made it
possible to observe roosting adults at various moult stages at any one time.
The acquisition of nuptial plumage and gonadal recrudescence stimulates the onset of
the courtship phase (Dunlop 1985a), where mature individuals attempt to form or re-establish
pair-bonds for the current breeding season. Mate fidelity has not been studied in Fairy Terns
but may occur in successful pairs, as has been documented for other larids such as Crested
Terns, Thalasseus bergii (e.g. Austin 1947; Coulson and Thomas 1983; Dunlop 1985a).
During the pair-formation phase, Fairy Terns engaged in social displays that were
characterised by aerial courtship flights, ground displays and the parading of fish by males
who, presumably attempted to demonstrate their quality to potential female mates. In other
terns, courtship feeding during the mate selection and pair formation periods provides the
female with an opportunity to evaluate the foraging ability of potential male mates (Nisbet
11
1973; Bried & Jouventin 2002), and, in Fairy Terns, is a necessary prelude to copulation.
However, once pair-bonds were formed, the function of courtship feeding, presumably
switches to supplementing nutrient and energy supply for the production of eggs (Lack 1968;
Nisbet 1973; Helfenstein et al. 2003).
Following the establishment of pair-bonds, pairs (usually in display flocks)
intermittently visited a colony area and prospected potential nesting sites. Established pairs
visited the preferred area for several days to weeks and females were often seen loafing and
being provisioned by males on shorelines close to the prospective colony site. Mating occurred
frequently throughout this period, but the exchange of fish between established pairs ceased to
be a necessary prelude to copulation. As egg-laying approached, males regularly provisioned
the female at the nest site, and she rarely left the nest territory, except to bathe and drink or
when threatened. The male continued to provision the female during the incubation period,
however, feeding rates decreased over time until egg-hatching when chick provisioning
commenced. Presumably, female provisioning during the incubation period helps to maintain
the pair bond and demonstrates the males’ potential to care for their brood (González-Solı́s et
al. 2001).
Extra-pair mating
Occasionally, extra-pair copulation and multiple female mating behaviour occurred in
and around Fairy Tern colonies, including among females with completed clutches (Supp. 2).
In all observed cases of extra-pair copulation, the exchange of fish was an essential prelude to
copulation, although, sometimes females stole the fish and flew away before the mating routine
was complete. In most cases, soliciting males were chased out of the territory and males may
guard female mates from single males (Supp. 2).
12
Colony formation, egg-laying and incubation
Colony site selection
Before the commencement of breeding, Fairy Terns actively prospected areas of
suitable nesting habitat that may be used as potential colony sites (Fig. 2). Environmental and
social cues likely played an important role in the selection of colony sites, with food
availability, habitat stability, vegetation characteristics, the presence of predators and previous
breeding success being some of the potential stimuli influencing site selection (McNicholl
1975; Gochfeld & Burger 1992; Boulinier et al. 1996; Feare et al. 1997; Paiva et al. 2006). Pre-
nuptial, immature and post-nuptial birds (possibly failed breeders) were frequently observed at
colony sites during the breeding season, which likely provide an opportunity for information
gathering and social learning.
Colony sites transmit information about habitat suitability and the potential for breeding
success; proximate cues for future settling decisions (Reed & Dobson 1993; Boulinier et al.
1996; Danchin et al. 1998). Yet Fairy Terns commonly shift colony sites between breeding
attempts. Even when colony sites have been highly successful or remain stable, sites may be
abandoned after one or more seasons (Gochfeld & Burger 1992; Friesen et al. 2017; Dunlop
2018). This shifting of colony sites may function to reduce predation risk, as the location of
colonies, presumably becomes predictable to predators over time. This life history trait makes
the development of Fairy Tern management strategies particularly challenging as site selection
is not predictable and subject to inter-annual change.
13
Figure 2. Summary of the breeding cycle and the factors influencing colony site selection and nesting success in the
Australian Fairy Tern, Sternula nereis nereis (migratory sub-population) in Western Australia. Components
highlighted in light green denote social paring and colony formation behaviour; components in blue represent the
influence of food availability on the breeding cycle and those in yellow represent the habitat. Fairy Tern sketches by
D. Chapman (2018).
The spatial and temporal availability of prey within close proximity of suitable nesting
habitat is likely to be a primary factor influencing colony site selection (Fig. 2, Veen 1977;
Dunlop 1987; Monaghan et al. 1989) since Fairy Terns are typically ‘single prey loaders’
(sensu Houston & McNamara 1985) and limited to carrying a single fish, held crosswise, in
their bill at any one time. This, undoubtedly, serves to maximise net energy gain by reducing
energy expenditure during foraging excursions.
Extensive observations of behaviour during the early colony formation period showed
that, initially, attachment to potential colony sites was low and prospecting bouts were limited
to a few birds flying over the area for brief periods before dispersing, and prospecting occurred
mainly in the few hours following sunrise. Over time, birds began alighting on the site and
engaged in site-attachment activities such as constructing nest scrapes and territory
establishment. The sites were abandoned by late morning or the afternoon, which is consistent
with the behaviour of other terns, including Common Terns, Sterna hirundo, Crested Terns,
14
and Sandwich Terns, Thalasseus sandvicensis (Veen 1977; Dunlop 1987; Gochfeld and Burger
1992), although, potential sites may be visited by single pairs or small groups at various times
throughout the day or night (Greenwell et al. 2019b). Multiple scrapes appear to be made before
a final nest site is selected, and the coming and going from the site may continue for several
days to weeks before any eggs are laid.
Nest construction, dispersion and habitat selection
Fairy Terns typically nested in open, sandy areas that were generally free from plants
or covered with a short, open herb or grassland. At Point Walter, nests were constructed among
native couch grass, Sporobolus virginicus and naturalised Atriplex prostrata, probably due to
a lack of sandy, vegetation-free areas above the high tide mark. The vegetation was flattened
and removed by the terns and where possible, eggs were laid onto the underlying sand.
Nest construction involved the excavation of a shallow scrape or depression into the
sand (~ 2-3 cm deep), and both members of the pair actively scraped the nest. Terns rearranged
shell material throughout the incubation period, and the incorporation of shell in and around
the nest probably plays an important role in supporting egg and chick crypsis, favouring
survival. While it’s possible that shell rearrangement behaviour may be a displacement activity
used to reduce stress (Clinning 1975), such behavioural traits are likely to be adaptive and
increase fitness within the population.
Colony development
At North Fremantle, colony growth was mapped over 6 weeks in 2018/19 (Fig. 3). The
colony expanded from a central position, termed the nucleus, which probably consisted of
older, more experienced birds (Nisbet et al. 1984). Over ¾ (77%) of all nests (n = 220) were
spaced < 1.0 m apart (median = 0.72 m; mean 1 SE = 0.89 0.05 m) but the distance from
nearest neighbour ranged from 0.30 to 5.69 m (Fig. 3). The median nest spacing generally
15
increased over time, from 0.89 m for nests established in week one to 1.04 m in week three and
2.06 m in week six, although, infilling caused a slight fluctuation in the median nest distance
in weeks two and five (see Shell cover preferences for detail).
Figure 3. Time series map showing colony development (by week) of an Australian Fairy Tern colony, Sternula
nereis nereis, at North Fremantle, Western Australia. Points indicate individual nests, colour-coded by the week
in which they were established. * denotes the location of the first nest to be established.
Pairs established and vigorously defended breeding territories, and social interactions
with neighbouring terns probably influenced nest spacing and the positioning of individual
nests within the colony (McKinney 1965; Temeles 1994; Greenwell, C.N. pers. obs.). Habitat
quality and complexity and the degree of familiarity a territory owner has with a neighbour are
likely to be important factors influencing nest spacing, with nests arranged at distances that
minimise predation risk and disruption (Lack 1968; Perrins & Birkhead 1983; Mackin 2005).
Group adherence, where members from a colony or sub-colony choose to remain
together over multiple breeding seasons, is important among some larids, and suspected in
Fairy Terns, and is thought to coordinate the re-establishment of colonies or sub-colonies in
new locations (Austin 1951; McNicholl 1975; Palestis 2014; Dunlop & Greenwell 2020).
Through processes of social attraction, discrete clusters of individuals with synchronized
reproductive periodicity were formed (Fig. 3, Supp. 2). Group adherence may play an
important role in the formation of these clusters and likely functions to enhance reproductive
success by increasing nest aggregation and synchrony (Austin 1951; McNicholl 1975; Veen
1977; Hernández-Matías et al. 2003), whilst simultaneously reducing intraspecific agonistic
16
behaviour and the requirement for territorial defence (McNicholl 1975). Synchronous hatching
in Fairy Terns is likely to increase chick survival through predator swamping, predator
confusion, and collective group defence behaviour (Lack 1968; Hamilton 1971; Estes 1976;
Ims 1990). Synchrony in nesting may also reduce predation risk to adults by reducing the time
a colony remains visible to predators (Atwood 1986; Greenwell et al. 2019b).
Shell cover preferences
In 2018/19, the percentage beach shell cover surrounding 112 nests was estimated at
the North Fremantle colony. Nest sites established within 7 days of the first egg being laid had
the highest mean shell cover ( 1 SE) at 73.5% 4.5% (Table 1). Shell cover reduced over
time to an average of 58.2% 7.9% for nest sites established more than 21 days after the first
egg was laid (Table 1). There was a significant relationship between percentage shell cover
and the date of nest establishment, with evidence of a decreasing trend in percentage cover
with time (RSE = 0.377, df = 112, p = 0.045). In addition, the average shell cover surrounding
the selected nest sites (n = 114, 64.9% 2.8%) was higher than randomly sampled sites (n =
44, 53.7% 4.4%; t = 12.34, df = 43, p < 0.0001). These findings suggest that birds breeding
early in the season select the best territories (Montevecchi 1978; Cody 1985) with high
percentage shell cover. Through processes of social facilitation, birds establishing nests were
attracted to other birds in the same reproductive state, causing some in-filling, yet, territories
comprising higher shell cover continued to be selected over time, when available and shell
rearrangement may be undertaken (Table 1, Fig 3).
Table 1. Mean shell cover (%, ± 1 standard error [SE]) within the nest territories of Australian Fairy Tern, Sternula
nereis nereis, at a colony in North Fremantle, Western Australia. Nest establishment = days after the first nest
was established.
Nest Establishment
No.
Territories
Mean Shell
Cover %
± 1 SE
< 8 days
37
73.5
4.5
8 -14 days
35
66.4
4.5
15-21 days
20
61.3
6.1
> 21 days
22
58.2
7.9
17
Egg-laying and incubation
Fairy Terns typically laid clutches of one to two eggs. Clutches of three eggs were rare
but may be produced by the more experienced pairs during seasons of high food abundance
(Coulson 1966; Nisbet 1973; Perrins & Birkhead 1983), female-female pairs, or may be the
result of egg-dumping (Shealer & Zurovchak 1995; Nisbet & Hatch 2008). During the 2018/19
season, no triple egg clutches were recorded at North Fremantle. However, 2.7% (n = 3) of
clutches at Mandurah consisted of three eggs, and in one clutch, all three eggs were observed
to hatch.
Of 220 nests recorded at North Fremantle in 2018/19, the first 92 to be established were
monitored twice each day. Of those, 82 (89%) hatched chicks. The incubation period ranged
from 17-26 days, with a median and mean of 21 days ( 1 SE = 0.17, Fig. 4). In multiple-egg
clutches, eggs hatched at intervals of zero to three days apart (mean = 1.06 0.02, n = 19). Six
nests (7%) were abandoned, while the egg did not hatch at three nests (3%). In all cases where
eggs did not hatch, pairs incubated for periods in excess of the observed maximum period for
successful clutches (≥ 32 days), suggesting that these eggs were infertile. During the 2019/20
breeding season on Penguin Island, multiple egg clutches (n = 8) hatched one to three days
apart (mean = 1.5 0.09).
Figure 4. Distribution of nest incubation periods of the Australia Fairy Tern, Sternula nereis nereis, at a colony
in North Fremantle, Western Australia in 2018/19. Note that the incubation period is restricted to single egg
clutch or the first egg to hatch within a clutch.
0
5
10
15
20
25
30
35
17 18 19 20 21 22 23 24 25 26
Number of Nests
Incubation Period (days)
18
The incubation period of Fairy Terns at North Fremantle was consistent with that of
other small terns. For example, incubation in the Damara Tern, Sternula balaenarum, ranges
from 17-22 days (Clinning 1975; Simmons & Braine 1994); Least Tern, Sternula antillarum,
17-24 days (U.S. Fish and Wildlife Service 1990), while incubation in the New Zealand Fairy
Tern, ranges from 22-25 days (average = 23 days, Parrish and Pulham 1995). Higher ambient
temperatures in Western Australia compared to New Zealand may explain the shorter average
incubation period between the two subspecies. Variability in the incubation period is likely
related to the total time adults spend thermoregulating their eggs, ambient air temperatures,
levels of colony disturbance and the time at which incubation becomes continuous (Kendeigh
1940; Nisbet 1975; Nisbet & Cohen 1975; Pettingill 1985).
Asynchronous breeding cycles among Fairy Terns can result in a protracted egg-laying
period, particularly at large colonies and may reflect age/experience of breeding birds, with
older birds generally laying earlier than later birds (Coulson & White 1958; Coulson 1966). In
2018/19 at the North Fremantle colony, eggs were laid over > 7 weeks from early-December
to mid-January (Fig. 5a). Two main peaks were observed, the first in early December and the
second in late December, with the large majority of nests (84%) being established by 29
December 2018, i.e. 4 weeks after the first egg was laid (Fig. 5a, b). These peaks probably
represent differing gonadal periodicities in the breeding cycle of Fairy Terns (Dunlop 1985b;
Dunlop & Jenkins 1992).
19
(a)
(b)
Figure 5. Australian Fairy Tern, Sternula nereis nereis, (a) egg-laying dates and (b) cumulative nest,
numbers, shown at three-day intervals for a North Fremantle colony in Western Australia during the
2018/19 breeding season.
Incubation shift duration
Incubation shift duration during the day for the 11 nests was highly variable, ranging
from 0.03 to 7.1 hr (mean = 1.27 ± 6.11 hr). Incubation duties were shared almost equally
between the sexes. The average daytime incubation period for males (mean = 1.31 ± 0.10 hr;
median = 1.10 hr) was slightly longer than for females (mean = 1.2 ± 0.07; median = 1.14 hr).
These findings are consistent with incubation periods of the Sandwich Tern but contrast with
the Little Tern and Common Tern, where females typically spend ~ 25% longer incubating
0
5
10
15
20
25
30
35
2-Dec
5-Dec
8-Dec
11-Dec
14-Dec
17-Dec
20-Dec
23-Dec
26-Dec
29-Dec
1-Jan
4-Jan
7-Jan
10-Jan
13-Jan
16-Jan
Number of Nests
0
20
40
60
80
100
2-Dec
5-Dec
8-Dec
11-Dec
14-Dec
17-Dec
20-Dec
23-Dec
26-Dec
29-Dec
1-Jan
4-Jan
7-Jan
10-Jan
13-Jan
16-Jan
Cumulative Number of Nests (%)
Laying Date
20
than males (Fasola & Saino 1995). Females were present on the nests at sunrise at the
commencement video recording periods for 13 of the 17 observations, which may indicate that
females incubate overnight, with a shift change occurring soon after males return from foraging
trips.
A best subset selection approach, using 200 repetitions of 10-fold cross-validation for
linear models of log-transformed incubation shift duration on all possible combinations of
Bureau of Meteorology weather for North Fremantle and Mandurah, suggested that a
combination of highest and lowest wind speeds, highest and lowest air temperatures as well as
relative humidity provided the best set of predictors of incubation shift duration (average mean
squared error of = 0.56). Incubation periods are likely to increase during stronger winds
because of the greater time it takes to locate and successfully capture prey. The requirement
for adults to shelter their brood during extreme weather conditions, such as during the heat of
the day, may reduce the shift time between members of a pair. Other factors not measured
during this study, such as prey availability, the distance from foraging grounds and individual
fishing efficiency, may also be important predictors of incubation duration.
Post-hatching behaviour
Parental care
Adults attentively brooded chicks for two to three days following hatching, although it
was not uncommon for chicks to be left unattended for short periods while adults (usually the
male) undertook foraging excursions. For example, at Rous Head on 26 December 2019, a
three-day-old chick was left unattended on six occasions for ~ 3 – 7 min before adults returned
with a fish and fed the chick. Alternative scrapes were routinely constructed a short distance
away from the original nest site and chicks were lured between scrapes using contact calls. In
one case, an adult was observed moving its chick to an alternative scrape within several hours
21
of hatching, i.e. while the chick was slightly wet. As chicks became more mobile, adults often
moved their brood to the periphery of the colony and established scrapes near vegetation, rocks
or other features that provided shelter and/or enhanced chick crypsis. Movement to the outskirts
of the colony probably also reduced competition for food, i.e. kleptoparasitism and the potential
for alloparental care, but may also provide relief from ectoparasites. Protection and defensive
behaviour among Fairy Terns was consistent with that of Little Terns, as described by Davies
(1981).
Sunrise to sunset observations from two nests during the chick brooding phase indicate
that males may provision chicks more frequently than females, as is the case with other larids
and surface-nesting colonial seabirds (e.g. Little Tern, Common Tern, Fasola and Saino 1995).
In North Fremantle, of 18 fish returned to nest 10B6, the male delivered 12, compared to the
female who returned 6 fish in 2018. At Mandurah, of 17 fish returned to nest CLA1, 14 and 3
fish were delivered by the male and female, respectively. Further research into diet and food
intake is required to understand parental roles and how provisioning may influence growth and
survival.
Intraspecific chick aggression and vocal recognition
Intraspecific aggression towards the young from other nests was commonplace within
Fairy Tern colonies, consistent with other terns (e.g. Feare 1976; Saino et al. 1994; Ramos
2003; Cabot and Nisbet 2013). Investing resources and care into non-genetically related
offspring has the potential to reduce individual fitness, particularly when it is carried out at the
expense of ones’ own progeny (Saino et al. 1994; Cabot & Nisbet 2013), and appears to occur
for Fairy Terns (see Supp. 2). Therefore, the evolution of behaviours that favour parent-chick
recognition and, conversely, prevent alloparental care should be promoted by natural selection
(Saino et al. 1994; Saino & Fasola 2010).
22
While independent recognition experiments were not conducted, extensive field
observations suggest that individual recognition of offspring takes several days to develop after
hatching, although, chicks may be able to recognise the vocalisations of their parents within
hours of hatching (Supp. 2). During the early post-hatching period, strong nest site attachment
and the visual stimuli provided by the presence of a chick within the nest appear to be the
primary cues influencing parental behaviour, as has been experimentally-tested in Caspian
Terns, Hydroprogne caspia (Shugart 1978). Like Caspian Terns, adult Fairy Terns appear to
develop recognition of their brood within 2-3 days of hatching, when the motivational stimuli
influencing parental responses switches, independently of nest location (Shugart 1978).
Before the development of individual recognition, there is the potential for vagile,
genetically unrelated chicks to be temporarily brooded or adopted by neighbouring terns
(alloparental care). This is due to an absence of physical barriers preventing chick exchange
within closely packed colonies and adults displaying extremely strong nest site attachment in
the early post-hatching period (Supp. 2). Nest site attachment appears to be so strong in the
first few days following hatching that vagile chicks who wander into nest scrapes are not
expelled, resulting in temporary or permanent alloparental care (Supp. 2). However, unattended
chicks, or those that wandered too close to neighbouring terns, were frequently picked up and
dumped in other parts, or on the outskirts, of the colony.
Multiple instances of alloparental care were recorded within Fairy Tern colonies during
the 2018/19 and 2019/20 breeding seasons (Supp. 2). On at least two occasions, unattended
vagile chicks were brooded by neighbours, but chicks subsequently returned to their birth-nest
on hearing parental contact calls. In 2018/19, a vagile chick was adopted by a non-related tern
at the North Fremantle colony, with chick acceptance into the nest captured on video (Supp.
2).
23
Time to fledging and post-fledging dispersal
Before fledging, chicks exercised their flight muscles and gained lift by jumping and
flapping into the wind. During strong winds, ~ 19-20-day old chicks became airborne for
several seconds but were not controlled in their flight. At Penguin Island, fledging occurred at
21-23 days (mean = 22 ± 0.21 d) (Supp. Table 2). Soon after fledging, chicks made short flights
to and from the colony but remained nearby whilst adults were out foraging. Within 8 days of
fledging, all banded juveniles had left the Penguin Island colony. One juvenile was sighted ~
35 km south on the Peel-Harvey Estuary ~ 24 hours after it was last sighted on Penguin Island.
Over a period of 8 weeks, all 10 surviving banded juveniles were resighted on sandbars and
spits within the Peel-Harvey Estuary (Fig. 1).
At the end of the breeding season, large numbers of Fairy Terns were observed staging
at Rottnest Island and on the Peel Harvey Estuary (Dunlop & Greenwell 2020, Fig. 1). These
areas appear to be important foraging sites for juvenile terns, where they hone their fishing
skills and feed, supplemented by adults, before making their journey north to their wintering
grounds at the Houtman Abrolhos (Dunlop & Greenwell 2020).
Nesting Success
Nesting success, as estimated by the number of fledglings produced, varied greatly
among colony sites. In 2018/19, ≥ 0.74 fledglings per pair were estimated at North Fremantle,
while at Mandurah, few birds fledged following depredation by a free-roaming cat, Felis catus,
and, subsequently, a Nankeen Kestrel, Falco cenchroides (Greenwell et al. 2019a). At Penguin
Island in 2019/20, nesting success was estimated at 0.33 fledglings per pair, following heavy
depredation by Silver Gulls, Chroicocephalus novaehollandiae, and a juvenile Crested Tern.
At Point Walter in 2018/19, suspected Red Fox, Vulpes vulpes, depredation appears to have
resulted in partial colony failure with an estimated ~ 0.25 fledglings per pair being produced
24
(fox prints detected, and damaged eggs and chicks found). In 2019/20, breeding success at
Point Walter was much higher (≥1 fledgling per pair), but to minimise disturbance at the site,
numbers were not recorded precisely.
Conservation implications
This research reveals important new insights into the breeding ecology, behaviour and
habitat preferences of Fairy Terns. Understanding animal behavioural ecology and natural
history is fundamental to conservation biology – informing approaches to management,
predicting and planning for environmental change, and understanding the resource needs of
those species requiring direct management intervention (Sutherland 1998; Berger-Tal & Saltz
2016). For conservation-dependent species, such as the Fairy Tern, the determinants of habitat
quality and the biological suitability of dedicated breeding sites must be given careful
consideration to ensure positive outcomes for breeding populations (Nisbet & Spendelow
1999). Ongoing threats and disturbance of breeding colonies highlight the need for direct
management interventions to be undertaken in many locations across their breeding range,
particularly those subject to high human activity (Ferreira et al. 2005; Baling et al. 2009;
Department of the Environment 2011; Commonwealth of Australia 2019; Dunlop & Greenwell
2020). Engineered colony sites (managed sites) provide an opportunity to overcome a lack of
natural nesting sites associated with coastal development, high levels of human-induced-
disturbance and increased risk of site inundation due to altered sea-level and climate change
(Threatened Species Scientific Committee 2011; Garnett et al. 2013).
Through processes of social facilitation, there is a tendency for conspecifics to be
attracted to established colonies as they come into reproductive condition. While social
facilitation results in an increase in colony size, the distance between neighbours may also
increase with time. Important consideration must be given by land and wildlife managers to
25
ensure that temporary fencing, deployed to protect colonies, is large enough to allow for colony
growth as the disturbance of peripheral nests may result in their abandonment. For managed
sites, providing large areas of substrate with high percentage shell cover may help ensure that
optimal habitat remains available for the duration of the nesting period.
Two main periods of high vulnerability have been identified in the reproductive season
for Fairy Terns and must be given special consideration when undertaking research and
monitoring activities. These include the early nesting period, when colonial surface-nesting
birds display low nest site tenacity (Safina & Burger 1983; Nisbet 2000), and the first few days
following chick-hatching before strong vocal recognition exists between an adult and its
offspring.
Finally, the tendency for Fairy Terns to shift colony sites from one breeding attempt to
the next, either within a season or between years, makes management challenging. However,
this feature of the life cycle is critically important for land managers to recognise and take into
account when preparing known breeding areas ahead of the breeding season (e.g. habitat
management, predator control, temporary fencing). Thus, multiple managed sites must be
considered and prepared, irrespective of whether the birds attempted to nest in the previous
season or not, to ensure the best possible chance of reproductive success each breeding season.
Endnote: Taxonomy in this study is as per the International Ornithological Congress
(IOC) world list. Note that "The Australian Fairy Tern Recovery Team does not accept a
separate sub-species in Western Australia. Recent investigations into the morphology and
genetics of Sternula nereis in Australia by L. Christidis, suggests that the DNA data is fully
consistent with the morphology in that S. n. hornii is not different to the nominate. This
information has been accepted by BirdLife International and will lead to changes in the
Handbook of the Birds of the World (Stephen Garnett, pers. comm.).” Therefore, the
correct nomenclature for the Australian Fairy Tern is Sternula nereis nereis.
26
Acknowledgements
We acknowledge the traditional owners of the Whadjuk and Bindjareb country, whose lands
we have had the great privilege of conducting this research upon. Sincere thanks to Brett
Newmarch and Merryn Pryor for their contributions to colony monitoring, data collection and
field support. We are grateful to Tegan Douglas for her technical advice on temporary colour
marking techniques. Special thanks are due to Adam van der Beeke and the team at Fremantle
Ports for their ongoing support of this research. To the Parks and Wildlife Service staff from
the Metropolitan Marine, Riverparks Unit and Swan Region, thank you for your in-kind
support and accommodation on Penguin Island, assistance and enthusiasm in supporting
research activities and colony monitoring at Point Walter and Penguin Island, where much of
this research was based. Thank you to Perth Wildlife Encounters for their generous in-kind
support, ferrying researchers to and from the Penguin Island during the breeding season.
Thanks are due to Phil Auty, Cherilyn Corker, Maggie Duggan, Sharon Manson, Fiona
O’Sullivan, Jeremy Ringma and Georgina Steytler for their support on Penguin Island. We are
grateful to Natalie Goddard and members of the Western Australian Fairy Tern Network for
providing valuable sight-recapture information on the fledglings banded as part of this
research. The banding study was conducted in accordance with Australian Bird and Bat
Banding Scheme approvals. Financial support was provided by Murdoch University,
Fremantle Ports, Stuart Leslie Bird Award & BirdLife Australia, and Holsworth Wildlife
Research Endowment & Ecological Society of Australia. We also thank the reviewers,
including Dr Maggie Watson, who provided constructive feedback to improve this manuscript.
27
References
Atwood JL. 1986. Delayed nocturnal occupation of breeding colonies by Least Terns (Sterna
antillarum). Auk 1:242–244. Estes & Lauriat.
Austin OL. 1947. A study of the mating of the Common Tern (Sterna hirundo). Bird-Banding
18:1.
Austin OL. 1951. Group adherence in the Common Tern. Bird-Banding 22:1–15.
Baling M, Brunton D. 2005. Conservation genetics of the New Zealand Fairy Tern. Auckland
UniServices Limited. Auckland, New Zealand.
Baling M, Jeffries D, Barré N, Brunton DH. 2009. A survey of Fairy Tern (Sterna nereis)
breeding colonies in the Southern Lagoon, New Caledonia. Emu 109:57–61.
Berger-Tal O, Saltz D. 2016. Conservation behavior: applying behavioral ecology to wildlife
conservation and management. Page (Berger-Tal O, Saltz D, editors). Cambridge
University Press, United Kingdom.
BirdLife International. 2018. Fairy Tern, Sternula nereis. Available from
http://dx.doi.org/10.2305/IUCN.UK.2018-2.RLTS.T22694691A132568135.en.
(accessed August 6, 2019).
Boulinier T, Danchin E, Monnat J-Y, Doutrelant C, Cadiou B. 1996. Timing of prospecting
and the value of information in a colonial breeding bird. Journal of Avian Biology 27:252–
256.
Bried J, Jouventin P. 2002. Site and mate choice in seabirds: an evolutionary approach. Pages
263–306 in E. A. Schreiber and J. Burger, editors. Biology of Marine Birds. CRC Press,
Boca Raton, London, New York, Washington.
Cabot D, Nisbet I. 2013. Terns. Collins, London.
Canty A, Ripley B. 2020. boot: Bootstrap R (S-Plus) functions. R package version 1.2-43.
Available at https://cran.r-project.org/web/packages/boot/index.html
Clinning CF. 1975. The biology and conservation of the Damara tern in South West Africa.
Madoqua 1978:31–39.
Cody ML. 1985. Habitat selection in birds. Academic Press.
Commonwealth of Australia. 2019. Draft National Recovery Plan for the Australian Fairy Tern
(Sternula nereis nereis). Available from
https://www.environment.gov.au/biodiversity/threatened/recovery-plans/comment/draft-
recovery-plan-australian-fairy-tern.
Coulson JC. 1966. The influence of the pair-bond and age on the breeding biology of the
Kittiwake Gull Rissa tridactyla. Animal Ecology 35:269–279.
Coulson JC, Thomas CS. 1983. Mate choice in the Kittiwake gull. Page 361— 376. in P.
Bateson, editor. Mate Choice. Cambridge University Press, Cambridge.
Coulson JC, White E. 1958. The effect of age on the breeding biology of the Kittiwake Rissa
tridactyla. Ibis 100:40–51.
Danchin E, Boulinier T, Massot M. 1998. Conspecific reproductive success and breeding
habitat selection: Implications for the study of coloniality. Ecology 79:2415–2428.
Davies S. 1981. Development and behaviour of Little Tern chicks (Sterna albifrons). British
Birds 74:291–298.
Department of the Environment. 2011. Approved conservation advice for Sternula nereis
nereis (Fairy Tern). Canberra. Available from
http://www.environment.gov.au/biodiversity/threatened/species/pubs/82950-
conservation-advice.pdf. (accessed June 15, 2020).
Department of the Environment. 2018. Sternula nereis nereis in species profile and threats
database. Available from http://www.environment.gov.au/cgi-
bin/sprat/public/publicspecies.pl?taxon_id=82950 (accessed August 14, 2019).
28
Dunlop J. 1987. Social behavior and colony formation in a population of Crested Terns, Sterna
bergii, in southwestern Australia. Wildlife Research 14:529. CSIRO Publishing.
Dunlop JN. 1985a. The relationship between moult and the reproductive cycle in a population
of crested terns, Sterna bergii Lichtenstein. Wildlife Research 12:487–494.
Dunlop JN. 1985b. Reproductive periodicity in a population of crested terns, Sterna bergii
lichtenstein, in South-Western Australia. Wildlife Research 12:95–102.
Dunlop JN. 2018. Fairy Tern (Sternula nereis) conservation in south-western Australia, 2nd
edition. Conservation Council of Western Australia, Perth, Western Australia.
Dunlop JN, Greenwell CN. 2020. Seasonal movements and meta-population structure of the
Australian Fairy Tern in Western Australia. Pacific Conservation Biology.
Dunlop JN, Jenkins J. 1992. Known-age birds at a subtropical breeding colony of the Bridled
Tern (Sterna anaethetus): A comparison with the Sooty Tern. Colonial Waterbirds 15:75.
Estes RD. 1976. The significance of breeding synchrony in the wildebeest. African Journal of
Ecology 14:135–152.
Fasola M, Saino N. 1995. Sex-biased parental-care allocation in three tern species (Laridae,
Aves). Canadian Journal of Zoology 73:1461–1467.
Feare CJ. 1976. The breeding of the Sooty tern Sterna fuscata in the Seychelles and the effects
of experimental removal of its eggs. Journal of Zoology 179:317–360.
Feare CJ, Gill EL, Carty P, Carty HE, Ayrton VJ. 1997. Habitat use by Seychelles Sooty Terns
Sterna fuscata and implications for colony management. Biological Conservation 81:69–
76.
Ferreira SM, Hansen KM, Parrish GR, Pierce RJ, Pulham GA, Taylor S. 2005. Conservation
of the endangered New Zealand Fairy Tern. Biological Conservation 125:345–354.
Foster MS. 1975. The Overlap of Molting and Breeding in Some Tropical Birds. The Condor.
Fox J, Weisberg S. 2019. car: An R companion to applied regression. Thousand Oaks CA:
Sage. Available from https://socialsciences.mcmaster.ca/jfox/Books/Companion/.
Friesen MR, Beggs JR, Gaskett AC. 2017. Sensory-based conservation of seabirds: a review
of management strategies and animal behaviours that facilitate success. Biological
Reviews 92:1769–1784.
Garnett S et al. 2013. Climate change adaptation strategies for Australian birds. Page National
Climate Change Adaption Research Faculty (NCCARF). National Climate Change
Adaptation Research Facility, Gold Coast.
Gochfeld M, Burger J. 1992. Family Sternidae (Terns). Pages 624–667 in J. del Hoyo, A.
Elliott, J. Sargatal, and J. Cabot, editors. Handbook of the birds of the world. Lynx
Edicions, Barcelona.
González-Solı́s J, Sokolov E, Becker PH. 2001. Courtship feedings, copulations and paternity
in common terns, Sterna hirundo. Animal Behaviour 61:1125–1132.
Greenwell CN, Calver MC, Loneragan NR. 2019a. Cat gets its tern: A case study of predation
on a threatened coastal seabird. Animals 9:445. Multidisciplinary Digital Publishing
Institute.
Greenwell CN, Dunlop JN, Loneragan NR. 2019b. Nest desertion: an anti-predator strategy of
the Australian Fairy Tern, Sternula nereis nereis. Marine Ornithology 47:197–201.
Greenwell CN, Woehler EJ, Paton D, Paton F, Dunlop JN, Menkhorst P, Carey M, Garnett ST.
2021. Australian Fairy Tern Sternula nereis nereis. In S. T. Garnett, editor. Action Plan
for Australian Birds 2020. CSIRO Publishing, Melbourne (In Press).
Hamilton WD. 1971. Geometry for the selfish herd. Journal of Theoretical Biology 31:295–
311. Academic Press.
Helfenstein F, Wagner RH, Danchin E, Rossi J-M. 2003. Functions of courtship feeding in
black-legged kittiwakes: natural and sexual selection. Animal Behaviour 65:1027–1033.
Hernández-Matías A, Jover L, Ruiz X. 2003. Predation on Common Tern eggs in relation to
29
sub-colony size, nest aggregation and breeding synchrony. Waterbirds 26:280–289.
Higgins P, Davies SJJF. 1996. Handbook of Australian, New Zealand and Antarctic birds.
Volume 3: snipe to pigeons. Oxford University Press, Melbourne, VIC; Aukland, NZ.
Hijmans RJ, van Etten J. 2012. raster: Geographic analysis and modeling with raster data.
Available at https://rspatial.org/raster Available at https://cran.r-
project.org/web/packages/doParallel/index.html
Houston AI, McNamara JM. 1985. A general theory of central place foraging for single-prey
loaders. Theoretical Population Biology 28:233–262.
Ims RA. 1990. On the adaptive value of reproductive synchrony as a predator-swamping
strategy. The American Naturalist 136:485–498.
Jeffries DS, Brunton DH. 2001. Attracting endangered species to “safe” habitats: Responses of
fairy terns to decoys. Animal Conservation 4:301–305.
Johnstone RE, Storr GE. 1998. Handbook of Western Australian Birds. Volume 1- non
passerines, Emu to Dollarbird. Page (Taylor DL, editor). Western Australian Museum,
Perth, Western Australia.
Kendeigh SC. 1940. Factors Affecting Length of Incubation. The Auk 57:499–513.
Lack D. 1968. Ecological adaptations for breeding in birds. Methuen, London.
Mackin WA. 2005. Neighbor-stranger discrimination in Audubon’s shearwater (Puffinus l.
lherminieri) explained by a “real enemy” effect. Behavioral Ecology and Sociobiology
59:326–332.
McKinney F. 1965. Spacing and chasing in breeding ducks. Wildfowl 16:92–106.
McNicholl MK. 1975. Larid site tenacity and group adherence in relation to habitat. The Auk
92:98–104.
Microsoft Corporation, Weston S. 2019. doParallel: Foreach Parallel Adaptor for the “parallel”
Package.
Microsoft Corporation, Weston S. 2020. foreach: Provides foreach looping construct.
Monaghan P, Uttley JD, Burns MD, Thaine C, Blackwood J. 1989. The relationship between
food supply , reproductive effort and breeding success in Arctic Terns Sterna paradisaea.
Journal of Animal Ecology 1:261–274.
Montevecchi WA. 1978. Nest site selection and its survival value among laughing gulls.
Behavioral Ecology and Sociobiology 4:143–161.
Nisbet ICT. 1973. Courtship-feeding, egg-size and breeding success in common terns. Nature
241:141–142.
Nisbet ICT. 1975. Selective effects of predation in a tern colony. The Condor 77:221–226.
Nisbet ICT. 2000. Disturbance, habituation and management of waterbird colonies. Waterbirds
23:312–332.
Nisbet ICT, Cohen ME. 1975. Asynchronoug hatching in Common and Roseate Terns, Sterna
hirundo and S. dougallii. Ibis 117:374–379.
Nisbet ICT, Spendelow JA. 1999. Contribution of research to management and recovery of the
roseate tern: review of a twelve-year project. Waterbirds 22:239–535.
Nisbet ICT, Winchell JM, Heise AE. 1984. Influence of age on the breeding biology of
Common Terns. Colonial Waterbirds 7:117–126.
Nisbet ICTT, Hatch JJ. 2008. Consequences of a female-biased sex-ratio in a socially
monogamous bird: female-female pairs in the Roseate Tern Sterna dougallii. Ibis
141:307–320. Wiley.
Paiva VH, Ramos JA, Catry T, Pedro P, Medeiros R, Palma J. 2006. Influence of environmental
factors and energetic value of food on Little Tern Sterna albifrons chick growth and food
delivery. Bird Study 53:1–11.
Palestis BG. 2014. The role of behavior in tern conservation. Current Zoology 60:500–514.
Parrish GR, Pulham GA. 1995. Observations on the breeding of the New Zealand Fairy Tern.
30
Tane 35:161–173.
Perrins CM, Birkhead TR. 1983. Avian ecology. Blackie & Son Limited, New York.
Pettingill OS. 1985. Ornithology in Laboratory and Field (Fifth Ed). Academic Press, Florida.
R Core Development Team. 2011. R: A Language and environment for statistical computing.
The R Foundation for Statistical Computing., Vienna, Austria. Available from
http://www.r-project.org/.
Ramos JA. 2003. Intraspecific aggression by Roseate Tern adults on chicks in a tropical colony.
Waterbirds 26:160–165.
Reed JM, Dobson AP. 1993. Behavioural constraints and conservation biology: Conspecific
attraction and recruitment. Trends in Ecology & Evolution 8:253–256. Elsevier Current
Trends.
Safina C, Burger J. 1983. Effects of human disturbance success in the Black Skimmer. The
Condor 85:164–171.
Saino N, Fasola M. 2010. The function of embryonic vocalization in the Little Tern (Sterna
albifrons). Ethology 102:265–271.
Saino N, Fasola M, Crocicchia E. 1994. Adoption behaviour in Little and Common Terns
(Aves; Sternidae): Chick benefits and parents’ fitness costs. Ethology 97:294–309.
Shealer DA, Zurovchak JG. 1995. Three extremely large clutches of roseate tern eggs in the
Caribbean. Colonial Waterbirds 18:105–107.
Shugart GW. 1978. The development of chick cecognition by adult Caspian Terns. Proceedings
of the Colonial Waterbird Group 1:110–117.
Simmons R, Braine S. 1994. Breeding, foraging, trapping and sexing of Damara Terns in the
Skeleton Coast Park, Namibia. Ostrich 65:264–273.
Sutherland WJ. 1998. The importance of behavioural studies in conservation biology. New
York 56:801–809.
Temeles EJ. 1994. The role of neighbours in territorial systems: When are they “dear enemies”?
Animal Behaviour 47:339–350.
Threatened Species Scientific Committee. 2011. Commonwealth Listing Advice on Sternula
nereis nereis. Canberra. Available from http://www.environment.gov.au/cgi-
bin/sprat/public/publicspecies.pl?taxon_id=82950 (accessed March 15, 2020).
U.S. Fish and Wildlife Service. 1990. Recovery plan for the interior population of Least Tern
Sternula antillarum. Twin Cities, Minnesota.
Veen J. 1977. Functional and causal aspects of nest distribution in colonies of the Sandwich
Tern (Sterna sandvicencis Lath.). Behaviour. Supplement:1–201.
Wickham H. 2019. stringr: Simple, consistent wrappers for common string operations.
Available from https://cran.r-project.org/package=stringr.
Wickham H, Seidel D. 2019. scales: Scale functions for visualisation. Available from
https://cran.r-project.org/package=scales.
31
Supplementary 1.
Table 1. Observations of incubation shift duration of Fairy Tern, Sternula nereis nereis, at North Fremantle and
Mandurah, Western Australia during the 2018/19 breeding season. Note that the first and last observations for
each day were excluded to remove uncertainty surrounding the length of an incubation shift. Therefore, the total
number of shifts and total incubation time includes only the observations for which the entire incubation period
is known.
Site
Nest
Observations
(Days)
Total No.
Shifts
Total
Incubation
Time
North Fremantle
8B12
3
25
31:38
8B16
2
15
22:15
8B19
2
14
21:47
9B1
1
8
10:58
9B8
1
7
13:10
9B14
1
5
12:04
10B6
3
21
34:53
Mandurah
CL1A
1
12
11:07
CL1B
1
10
12:15
CL2A
1
8
09:07
CL3A
1
7
11:35
TOTAL
17
132
98:33
Table 2. Banding records showing time to fledging of Fairy Tern, Sternula nereis nereis chicks (n = 10), banded
at Penguin Island, Western Australia in January 2020. Fledging is defined as the development of controlled and
independent flight, i.e. not wind assisted. Mean time (± 1 SE) to fledging = 22 ± 0.21 days.
Band
Colour
Nest
Hatch Date
Fledging Date
Time to
Fledging
Pink
5
6/1/20
28/1/20
22
Green
3
6/1/20
28/1/20
22
White
8
6/1/20
28/1/20
22
Purple
3
7/1/20
29/1/20
22
Orange
9
7/1/20
28/1/20
21
Red
6
6/1/20
28/1/20
22
Light Green
4
6/1/20
29/1/20
23
Gold
16
9/1/20
30/1/20
21
Grey
4
9/1/20
1/2/20
23
Dark Blue
20
9/1/20
31/1/20
22
32
Supplementary 2 – Field Notes
Extra-pair mating
In Mandurah, during the 2018/19 season, a female, paired with a red colour-banded
male, engaged in extra-pair mating with an unbanded male. Following the exchange of fish and
apparent successful copulation, the female returned to her nest and resumed incubation.
Importantly, the incubation period of the two eggs at this nest was well advanced. Therefore,
it is unlikely that the extra-pair mating was of any benefit to the male. At least three instances
of consecutive mating with multiple mates have also been observed, despite the females having
completed egg-laying. Occasionally, females solicit or accept advances from extra-pair, but
before the male is able to copulate, she abruptly ends the mating attempt by stealing his fish
and chasing him out of the territory
Mate guarding
Mate competition occurs throughout the egg-laying and incubation period at Fairy Tern
colonies, with unpaired males commonly seeking female mates. Males land within the colony
parading fish, in an attempt to attract single females or promiscuously mate with paired
females. Soliciting males are usually chased out of the territory quickly. However, physical
guarding of a female was observed on one occasion.
During the 2018/19 season at North Fremantle, a paired male announced his return to
the colony with a fish by vocalising to his female mate. The female stepped away from the nest
and greeted the male who was subsequently courtship-fed. The male mate approached the
pair’s nest and commenced incubation. Simultaneously, a male competitor arrived in the
territory with a fish. Instead of the female driving him off, the female proceeded to beg for the
fish – she assumed a bent posture, began vocalising (female beg) and fluttering her wings. The
male competitor proceeded to parade around the female with the fish and approached from
33
behind, in an attempt to commence pre-copulatory displays. The male mate began calling and
displaying agonistically to the competing male and moved between the intruder and the female,
blocking his advances. The male mate then commenced pre-copulatory displays with the
female, which concluded with mounting and copulation. Immediately following copulation,
the female chased the intruding male from the territory. Subsequently, both members of the
pair resumed normal behaviour, with the female returning the nest and the male resting nearby.
Chick adoption or expulsion
At Point Walter on 18 December 2019, a 1-2-day old chick strayed near the nest of a
neighbouring adult while its parent was out foraging. Despite being attacked initially, the chick
clambered underneath the neighbour and was diligently brooded. On hearing the parent
vocalise as it returned to the colony with a fish, the chick reared its head from underneath the
broody neighbour and begged vigorously for the fish. The chick continued to react to its parents
contact calls and made numerous attempts to approach the parent, but on each attempt the
neighbouring tern used its bill to push the chick back down in its own nest scrape, which
contained at least one egg. The parent searched the immediate area and continued to vocalise
within its territory for more than 10 min before swallowing the fish and returning to its nest on
two occasions. After nearly two hours, the adult returned with a third fish. The chick managed
to escape from underneath the neighbour and clambered back towards the nest in direct
response to the parents’ contact calls. The chick remained within a short distance of the birth-
nest for 7 min before being offered another fish by the parent, at which time it clambered back
into its nest where it was brooded.
At North Fremantle on 2 January 2019, a ~2-day old chick was picked up by an adult
from a neighbouring nest (10A3) and dumped close to the nest of 11B7. The chick proceeded
to approach the nest of 11B7 and was initially attacked by the adult. However, the chick
34
scurried underneath the adult and was diligently brooded along with a biologically related? To
whom? chick and was, ultimately, adopted. The adopted chick was observed over several days
close to the nest cup and by 6 January 2019 was observed begging on hearing the parent return
to the colony site with fish. Earlier in the day, a second chick from 11B7 was carried away by
an unknown adult and dumped in the colony. Incidentally, the event was captured on video and
shows the siblings being tended to by the parents in the morning, the adoption event and the
adopted chick’s initial responses to adults returning with food. On the day of adoption, the
adopted chick does not recognise the calls of the foster parent, unlike the biologically related
chick.
At North Fremantle, a vagile ~ 10-day old chick was observed in the nest of 10C6 (6
January 2019), being brooded by the adult, whose biological chick had hatched the day before.
Similarly, on a hot (~34 ºC) afternoon, on Penguin Island (15 January 2020), a ~ 10-day old
chick vagile chick was observed in the nest of X10 with one of the chicks banded as part of
this study. The vagile chick was diligently brooded by the adult for several hours during the
hot afternoon while its parents were out foraging. The chick was observed moving back to its
territory on two occasions when the parents returned with fish, where it was fed.
